Brushless Dc Motor 00 Du Ru

8/12/2019 Brushless Dc Motor 00 Du Ru

http://slidepdf.com/reader/full/brushless-dc-motor-00-du-ru 1/112



•3003







NAVAL POSTGRADUATE SCHOOL

Monterey, California

THESISBRUSHLESS DC MOTORS, VELOCITY AND

CONTROL OF THE BRUSHLESS DC

POSITIONMOTOR

by

Nezih Y. Durusu

June 1986

Thesis Advisor: George J. Thaler

Approved for public release; distribution is

unl imited.

T230364





CLASSIFICATION OP THIS PAGE

REPORT DOCUMENTATION PAGE

SECURITY CLASSIFICATION lb. RESTRICTIVE MARKINGS

CLASSIFICATION AUTHORITY

/DOWNGRADING SCHEDULE

3 DISTRIBUTION /AVAILABILITY OF REPORT

Approved for public release;distribution is unlimited.

ORGANIZATION REPORT NUMBER(S) 5 MONITORING ORGANIZATION REPORT NUM8ER(S)

OF PERFORMING ORGANIZATION

Postgraduate School

6b OFFICE SYMBOL(If applicable)

Code 62

7a NAME OF MONITORING ORGANIZATION

Naval Postgraduate School

(City, State, and ZIP Code)

California 93943-5000

7b ADDRESS (C/ty, State, and ZIP Code)

Monterey, California 93943-5000

OF FUNDING /SPONSORING 8b OFFICE SYMBOL(If applicable)

9 PROCUREMENT INSTRUMENT IDENTIFICATION NUMBER

(City, State, and ZIP Code) 10 SOURCE OF FUNDING NUMBERS

PROGRAMELEMENT NO

PROJECTNO

TASKNO

WORK UNIT

ACCESSION NO

(Include Security Classification)

DC MOTORS, VELOCITY AND POSITION CONTROL OF THE BRUSHLESSMOTOR

AUTHOR(S)

Nfisih YOF REPORT

Thesis13b TIME COVERED

FROM TO

14 DATE OF REPORT (Year, Month, Day)

1986 June15 PAGE COUNT

103

EMENTARY NOTATION

COSATI CODES

GROUP SUB-GROUP

18 SUBJECT TERMS (Continue on reverse if necessary and identify by block number)

Brushless DC Motors, Velocity and PositionControl, Microprocessor Controller

{Continue on reverse if necessary and identify by block number)

feedback controller for the brushless DC motor was designedthe Hall effect sensors. In addition, the position control of the

DC motor was developed using an optical encoder to sense angulartion changes and a microprocessor to provide the desired position

A Pittman 5111 wdg #1 brushless DC motor was used for this study,design of the digital tachometer and pulse width modulator for velocity

and the design of the Z-80 based microprocessor controller anddesign are described in detail.

OF ABSTRACT

SAME AS RPT D DTIC USERS

21 ABSTRACT SECURITY CLASSIFICATION

UNCLASSIFIED

OF RESPONSIBLE INDIVIDUAL

J. Thaler22b TELEPHONE (Include Area Code)

408-646-213422c OFFICE SYMBOL

62Tr

1473. 84 mar 83 APR edition may be used until exhaustedAll other editions are obsolete

1

SECURITY CLASSIFICATION OF this PAGE

UNCLASSIFIED



Approved for public release; distribution is unlimited.

Brushless DC Motors, Velocity and Position Control of theBrushless DC Motor

by

Nezih Y. DurusuLieutenant Junior Grade, Turkish Navy

B.S. ,' Turkish Naval Academy, 1978

Submitted in partial fulfillment of the

requirements for the degree of

MASTER OF SCIENCE IN ELECTRICAL ENGINEERING

from the

NAVAL POSTGRADUATE SCHOOL

June 1986 ^-^



ABSTRACT

A velocity feedback controller for the brushiess DC

motor was designed using the Hall effect sensors. In

addition, the position control of the brusnless DC motor

was developed using an optical encoder to sense anguiar

position changes and a microprocessor to provide the desired

position control. A Pittman 5111 wdg #1 brushiess DC motor

was used for this study. The design of tne. digital tachometer

and pulse width modulator for velocity control and the

design of the Z-80 based microprocessor controller and

software design are described in detail.



TABLE OF CONTENTS

I. INTRODUCTION 11

II. CONSTRUCTION AND OPERATION OF BRUSHLESSDC MOTORS 15

A. CONSTRUCTION OF BRUSHLESS DC MOTORS 15

B. ELECTRONIC COMMUTATION AND DRIVE 17

C. FOUR-PHASE DELTA BRUSHLESS dC MOTOR 21

D. ADVANTAGES AND DISADVANTAGES OF THE

BRUSHLESS DC MOTOR 25

III. VELOCITY CONTROL OF THE 3RUSHLESS DC MOTOR . . . . 27

A. GENERAL 27

B. DESIGN OF THE DIGITAL TACHOMETER FOR

VELOCITY CONTROL SYSTEM 30

C. PULSE WIDTH MODULATOR 34

IV. SYSTEM TESTING AND DATA COLLECTION FOR

VELOCITY CONTROL SYSTEM 39

A. GENERAL 39

B. OPEN LOOP VELOCITY CONTROL 40

C. CLOSED LOOP VELOCITY CONTROL 43

D. TRANSFER FUNCTION MEASURE AND SIMULATIONSTUDIES 46

V. POSITION CONTROL OF THE DC MOTOR WITHMICROPROCESSOR CONTROL 58

A. GENERAL 53

3. MICROPROCESSOR CONTROL OF DC MOTORS 58

C. INCREMENTAL OPTICAL ENCODER 59

D. MICROCOMPUTER SYSTEM 62

E. SOFTWARE DESIGN 65



VI. SYSTEM TESTING AND DATA COLLECTION FOR

POSITION CONTROL SYSTEM 70

A. GENERAL 70

B. SYSTEM CALIBRATION 71

C. CLOSED LOOP POSITION CONTROL 71

VII. SUMMARY AND CONCLUSION 76

A. REMARKS AND CONCLUSIONS 76

B. RECOMMENDATIONS FOR FURTHER STUDIES 77

APPENDIX A: RATING AND SPECIFICATIONS FOR MOTOR ... .30

APPENDIX 3:

APPENDIX C:

APPENDIX D:

APPENDIX E:

THE DAC CALIBRATION FOR VELOCITY CONTROL

SYSTEM 31

ARTWORK OF THE DIGITAL TACHOMETER AND

PULSE WIDTH MODULATOR CIRCUIT 32

THE DAC CALIBRATION FOR POSITION CONTROLSYSTEM 84

CHARACTERISTICS OF THE OPTICALINCREMENTAL ENCODER 35

MC 6366 1B OPERATION AND PROGRAMMING . . .37

INTEL 8255A OPERATION AND PROGRAMMING . .39

MAIN PROGRAM 91

LIST OF REFERENCES 101

INITIAL DISTRIBUTION LIST 102

APPENDIX F

APPENDIX G

APPENDIX H



LIST OF TABLES

TABLE 4.1. FREQUENCY RESPONSE WITH MAGNITUDEAND PHASE 50

TABLE 6.1. COUNTS AND ANGULAR POSITIONS 74



LIST OF FIGURES

Figure 2.1. Cut-Away View of Conventional DC Motor

Assembly 15

Figure 2.2. Cut-Away View of Brushless DC .MotorAssembly 16

Figure 2.3- Essential Parts of a Conventional DC

Motor 13

Figure 2.4. Essential Parts of a Brushless DC Motor .19

Figure 2.5. A Three-Phase, Half Wave Brushless DC Motor

Controller 20

Figure 2.6. Two-Phase Brushless DC Motor 22

Figure 2.7. Four-Phase Commutation Circuit 24

Figure 2.8. Four-Phase Logic Control 25

Figure 3.1. Block Diagram of the Velocity ControlSystem 28

Figure 3.2. Position of the Hall Effect Sensors . . .29

Figure 3.3. Flux in the Air Gap and Output VoltageWave Forms for Hall Effect Sensors . . . .29

Figure 3.4. One Channel Output of Hall Sensor . . . .31

Figure 3.5. Circuit Diagram of Digital Tachometer . .32

Figure 3.6. Output of the Flip Flop 33

Figure 3.7. Output of the Multivibrator 33

Figure 3.8. Digital to Analog Conversion 34

Figure 3.9. Pulse Width Modulator 35

Figure 3.10. Error and Dither Signal 36

Figure 3.11. Circuit Diagram of the Pulse WidthModulator 37

Figure 3.12. Circuit Diagram of the Digital Tachometerand Pulse Width Modulator 38

7



Figure 4.1.

Figure 4.2.

Figure 4.3.

4.4.

Figure 4.5.

Figure

Figure

Figure

4.6.

4.7.

Figure 4.3.

Figure 4.9.

Figure 4.10.

Figure 4.11a

Figure 4.11b

Figure

Figure

Figure

4.12.

4.13.

4.14.

Figure

Figure

4.15.

4.16.

Figure 4.17.

Figure 5.1.

Block Diagram of the Velocity ControlSystem 39

Hall Sensor Output of the Motorfor 3750 RPM 41

Pulse Width Modulated Signal

for 3750 RPM 42

Hall Sensor Output of the Motorfor 3260 RPM 42

Pulse Width Modulated Signalfor 3260 RPM 43

Dither Signal 44

Hall Sensor Output of the Motor

for 1275 RPM 45

Pulse Width Modulated Signalfor 1275 RPM 45

Pulse Width Modulated Signal withExternal Force on the Shaft Motor . . . .46

Closed Loop System with SpectrumAnalyzer 47

Open Loop Frequency Response of the Systemwith Magnitude Curve 47

Open Loop Frequency Response of the Systemwith Phase Curve 43

Bode Plot of the System 53

Frequency Response of the System 54

Transient Response of the Systemfrom Computer Simulation 55

Open Loop Transient Response of the

System from the Strip Chart 56

Closed Loop Transient Response of the

System from the Strip Chart 56

Closed Loop Transient Response of the.

System from Storage Oscilloscope 57

Position Control System 59

3



Figure 5.2.

Figure 5.3.

Figure 5.4a

Figure 5.4b

Figure 5.5.

Figure 5.6.

Figure 5.7.

Figure 6.1.

Figure 6.2.

Figure 6.3.

Figure 6.4.

Figure 6.5.

Direction Sensor 60

Incremental Optical Encoder 60

Encoder Channel Outputs for CW Rotation .63

Encoder Channel Outputs for CCU Rotation .63

Microcomputer System 64

Circuit Diagram of the Microprocessor . .68

Flow Chart of the Main Program 69

CRT Terminal Menu for Program Options . .72

CRT Terminal Menu for Calibration of

System 73

CRT Terminal Menu for Position Control . .73

Block Diagram of the PositionControl System 75

Curve Following Block 75



ACKNOWLEDGMENT

I wish to express my deepest thanks to Dr. George

Thaler for nis confidence and guidance which contributed to

the completion of this thesis. I would also like to express

my gratitude to my wife, Candi, for her loving support as

well as her typing and editing abilities.

10



I. INTRODUCTION

In recent years, the brushless DC motor has found more

applications because of its many advantages. It offers long

operational life, it eliminates brushwear particles and arcing,

and it is adaptable to spacecraft requirements.

As more specialized needs become obvious, the versatility

of the brushless DC motor in applications to control systems

was discovered and developed.

The brushless DC motor is mainly an inside out version

of the conventional DC motor. The rotor consists of permanent

magnets and the windings are in the stator. Besides this,

the areas where the conventional DC motor and brushless DC

motor differ are in the commutation processes and the

amplifier design. The commutation of the conventional

DC motor is done by a mechanical commutator and brushes.

On the other hand, the commutation of the brushless DC

motor is performed by semiconductor switching elements,

usually transistors. The inductive switching energy is

dissipated through a diode path which allows the current to

decline in a controlled fashion.

The commutation sensor system for brushless DC motors

is required to control the logic functions of the controller

to maintain current to the proper coils in tne stator.

Hall effect sensors and optical incremental encoder sensors

11



are the most commonly used methods for the angular position

sensing system.

The Hall effect sensing system is based on sensors

which are usually placed in the stator structure to sense

the polarity and magnitude of the permanent magnet field in

the air gap.

The optical increment encoder provides a pulse for eacn

increment of angular resolution. It is most commonly a

combination of light-emitting diode (LED), rotating disk,

mask and phototransistor .

The Pittman 5111 Wdg #1 brushless DC motor and four-pnase

drives were used for this study. One motor had a Hall sensor

and another motor had a Hall sensor and an optical incremental

encoder as well.

The velocity control of the system was designed by using

the fact that the Hall sensor gives two pulses per revolution

for a four pole motor. By counting the intervals between

each revolution, the digital speed can be obtained. V/iuh

this idea in mind, a digital tachometer was designed. The

speed command was given by dip switches and converted to

the analog signal. The digital speed which was obtained

from the digital tachometer was converted to an analog

signal with a Digital to Analog Converter (DAC).

The Pittman four-phase drive accepts four inputs. Two

of them are the logic signals from the Hall effect sensors.

One of the inputs is the direction command. The other

12



input is used for on-off control of the motor which is a

convenient logic input to apply a pulse width modulation

signal for speed and tdrque control. Keeping this feature

in mind, the pulse width modulator was designed.

In recent years microprocessor systems have been useful

tools with many applications. These involve the use of the

brushless DC motor, and the microprocessor control of the

brushless DC motor. Of its many features one of the most

important is the ease with which a system can be modified to

perform new functions. This can be easily done by writing

a new software program. Assembly language or high level

languages such as Forth, Basic, Fortran, C, Pascal and Ada

can be used for programming and can be downloaded to the

EPROM.

The microprocessor controller was designed by using a

Z-80 control processor unit. Parallel interfacing was used

to communicate with the outside world (the CRT terminal and

pulse width modulator). Position commands were given from

the CRT terminal and the updated position of the motor was

observed from the terminal also.

In Chapter Two the brushless DC motor is compared with

conventional DC motors and drive circuits. The third chapter's

emphasis is on the velocity control of the brushless DC motor

and the building of the digital tachometer and pulse width

modulator. In Chapter Four testing and data collection of

the velocity control system are studied. The position

13



control of the system with the microprocessor controller is

discussed in Chapter Five. In Chapter Six testing and data

collection of the position control system are studied.

14



II. CONSTRUCTION AND OPERATION OF BRUSHLESS DC MOTORS

A. CONSTRUCTION OF BRUSHLESS DC MOTORS

Brushless DC motors, unlike conventional DC motors have

a permanent-magnet rotor and a multi-coil stator. It can be

said that the basic brushless DC motor is essentially an

inside out version of the conventional DC motor. A cut-away

of a conventional DC motor is shown in Figure 2.1 and an

equivalent version of a brushless DC motor is shown in Figure

2.2. Here we can see the permanent magnet rotor and a

multi-coil stator.

MAGNET

ROTOR

Figure 2.1. Cut-Away View of Conventional DC Motor Assembly.

15



STATOR

PERMANENT - MAGNETROTOR

Figure 2.2. Cut-Away View of Brushless DC Motor Assembly.

A significant difference can be seen in the winding and

magnet locations. The conventional DC motor has the active

conductors in the slots of the rotor, and in contrast, the

brushless DC motor has the active conductors in slots of the

stator. Since the windings are closer to the environment the

removal of the neat produced in the active windings is easier

in the brushless DC motor. The result is that the brushless

DC motor is a more stable mechanical device from a thermal

point of view.

Another basic difference from the conventional DC motor

is the commutation process. The commutation of the conven-

tional DC motor is done by a mechanical commutator and

brushes. The brushless DC motor, on the other hand, is

commutated electronically.

16



B. ELECTRONIC COMMUTATION AND DRIVE

In order to see the similarities and differences between

conventional and brushless DC motor systems, two sketches are

shown in Figures 2.3 and 2.4. In Figure 2.3 we have the

elements of DC motor and control. The connections between

the rotor windings and the commutator are shown. In Figure

2.4 the commutation control stage is different from tne

conventional DC motor. Slots, windings, magnetic poles, and

the electronic commutator work in such a way that the

direction of the rotation is controlled by the polarity of

the DC power supply. By an electronic commutator, the

current is switched from one coil group to the adjacent one

with a four section stator winding. Switching takes place

from one coil to the next four times per revolution for a

two pole motor. Since the switching transistors are already

in place in electronic commutation, pulse width modulation

can be applied to the logic circuit. The shaft position

sensor creates pulses to generate logic signals which

control the commutation of the windings.

One of the simple, three-phase brushless DC motor

circuits is shown in Figure 2.5. This is a

half-wave

control circuit with a conduction angle of 120°. As is

shown, each winding is used one third of the time and the

logic control of the system is not complicated. The speed

and torque output of the motor can be controlled by varying

the power supply voltage Vs

. In the lower part of the

17



diagram the same system can be seen in reversed torque.

The torque reversal in a conventional DC motor is achieved

by reversing the power supply voltage. In the brushiess DC

motor the same tning can be done by shifting all the logic

functions by 180°. This example illustrates one of the

basic differences between conventional and brushiess DC motors.

In the illustration of Figure 2.5, the inductive transient

current in each winding is ignored. Due to the voltage

produced by the stored

0RivE=»

C.PCblT

17

LOGIC

ORCuiT

CGMMANOSIGNAL

INPUT

POWER

AC OR

DC

POWER

Supply

LOGIC POWER

BRUSHES

COMMUTATOR

Figure 2.3. Essential Parts of a Conventional DC Motor

18



COMMUTATION CONTROLLER

STAGE

Figure 2.4. Essential Parts of a Brushless DC Motor.

energy in each winding, the circuit creates a reverse breakdown

voltage on each transistor. Since stored energy is low in

the low-power systems, such break-down conditions can be

tolerated. However, if any significant amounts of current

and voltage are handled in such a system, breakdown conditions

would cause damage to the semiconductor junctions. Therefore

other methods are used to maintain proper commutation of

the inductive energy in each winding. Figure 2.6 shows a

two-phase brushless DC motor using two power supplies +V

and -Vs

. We now have four power transistors and four

19



+ v

GO 120 130 240 300 360 60 120Sha It angle

POSI TIVE

TORQUE

Figure 2.5. A Three-Phase, Half Wave Brushless MotorController.

20



diodes. Each half of the circuit controls it's own winding,

and the two operate independently of each other.

The diagram shows the current response with respect to

rotor position and current versus time at a given shaft

velocity.

It can be seen that the current Iq^ has an exponential

initial increase to a steady state value which is maintained

until tne 90° position has been reached. Then Q1 is switched

to the off condition. The stored energy is dispelled through

the power supply by using diode D3, and an exponential decline

is shown in Ipo, when the current rise is now progressing in

Q2. Tnus there is a continuous torque production maintained

in the motor as one stage is turned off and the next is

turned on.

C. FOUR-PHASE DELTA BRUSHLESS DC MOTOR

A four-phase Delta motor from the Pittman Corporation (see

Appendix A) is used for the following experiments. A four

pole structure is used for this motor.

There are several reasons to fabricate the rotor as a four

pole structure:

i) Mechanical arc lengths of 60° per magnet segment yielda higher material utilization than 120° arc used for

a two pole structure and therefore lower cost.

ii) High performance magnetic materials do not accept radialmagnet paths and thus are not as efficient magneticallyif made in long arc lengths.

iii) The four-pole structure doubles the number of commutationcycles per mechanical revolution of the shaft.

21



V

.

1

02

CO

n «> nrjy^-,D

f

D4

D4

c o-^WJP-o

90 180 270 360 90 ISO

iiKirmi dcgrM*

1 • Transistor ON2 • Tranirstor OFF

Figure 2.6. Two-Phase Brushless DC Motor

22



A four-phase commutation circuit is shown in Figure 2.7.

The logic outputs from the sensors are connected to a

BCD to decimal decoder by using the A and B inputs.

The C input is used to control the rotation of the motor.

D input is used for on-off control of the decoder. The

D input of the decoder in the motor drive is a convenient

logic input to apply a pulse width modulation signal for

speed and/or torque control. More details will be discussed

in the later sections. The flux rotation is provided by

the on , off position of the transistors. When a transistor

is on, it creates current on the related windings. The

current passing through the transistor will create flux on

the related windings. The driver controls the s tat or

excitation. For the clock-wise (CW) direction of the flux

rotation, transistors Q1 and Qg are on. This means that D

phase will have positive voltage and B phase will nave

negative voltage. In the next step, Q2 and Q4 ' transistors

will be on. This will create positive voltage at the C

phase and negative voltage at the A phase. This will

continue in the order: transistors Q3 and Q1 ' on and tran-

sistors QU and Q2 ' on. To reverse the flux direction,

the operating program will be transistors Q 1 and Q 3 ' on,

transistors QU and Q2 ' on, transistors Q3 and Q1' on,

transistors Q2 and Q4 1 on. Shaft angle position, phase

voltage, and corresponding sensor signals are shown in

Figure 2.8.

23



3NI1 t?

01 3NH 8

-N/vSr

na3aoo3a

< CD O QNouo3yia

Sb0SN3S

319NV 13VHS

WMd;•

Figure 2.7 Four-Phase Commutation Circuit

24



CW TORQ.UE

Electricalangularposition

SENSOR

1

SENSOR

2

PHASEA B I

C

TRANON

VOL TRANON

VOL TRANON

VOL TRANON

VOL

0°1 0.4 + 02 —

90° 03 -t- Ol'—

130° 1 GL4-'— 02 +

270 1 103' 01 4-

COW TORQUE

I04-' — 02 -h

270 I1 03 4- Ol'

—

180° 1 04 -4- 02' —

90°i

03' —Ol -r

Figure 2.3. Four-Phase Logic Control.

D. ADVANTAGES AND DISADVANTAGES OF THE BRUSHLESS DC MOTOR

1 . Advantages

Brushless DC motors are more expensive for the same

horsepower rating than conventional DC motors, but they have

some advantages over DC commutator-brush motors:

a) The motor has a long life because it does not

have brushes.

b) Due to the elimination of brush arcing, there is

a reduction in electromagnetic interference.

c) There is a reduction in acoustic noise.

d) Little or no maintenance is required.

e) The motor permits a small signal control of

speed and on-off operation since tne powercircuitry is included as part of the brushless

DC motor.

25



f) When they are properly sealed, they are capableof operation in fluids or vapors.

2. Disadvantages

The following are important disadvantages of the

brushlessDC

motor:

a) The total size of the motor is bigger overallbecause of the additional space required for the

electronic devices.

b) Overall cost is higher compared to conventionalcommutator types of the same horsepower.

c) Choice is somewhat limited at present in stocksizes and horsepower rating, necessitating special orders for particular applications.

Even though the brushless DC motor has some dis-

advantages, developing electronic technologies and applications

in space and the military make it preferable to conventional

DC motors.

26



III. VELOCITY CONTROL OF THE BRUSHLESS DC MOTOR

A. General

Before studying the speed control of the brushless DC

motor, it will be helpful to study the components of the

system. The block diagram of the velocity control circuit

is shown in Figure 3.1.

The Hall effect sensing system is based on sensors which

are located adjacent to the end of the stator winding to

sense the polarity and magnitude ofthe

permanent magnet

field in the gap. The position of the Hall sensors are

shown in Figure 3.2. The Hall effect device is made of two

sensors which are placed 90 electrical degrees apart to

sense the rotational position of the rotor relative- to the

stator coil groups. The flux in the gap between the rotor

and 'stator and the output of each sensor is shown in Figure

3.3. As can be seen in Figure 3.3, the output of each Hall

sensor switches from logic high to logic low when the

sensed rotor flux passes through zero. The output is high

for a north magnetic pole and low for a south pole (or vice

versa if Hall sensors are reverse mounted). [Ref 4]

The two rotor position signals are decoded by digital logic

gates in the motor drive to give a four phase output which

controls 8 power transistors in such a way that sequential

switching from one stator coil to the next occurs at intervals

of 90° mechanical rotor rotation. Both the outputs of the

27



DIRECTION

vref

tach

D/A CONVERTER

DIGITAL

TACHOMETER

PWM PWM

ii A

D C

MOTOR

ADRIVE

B [76-6-1]

0A 0BHALL

SENSORMOTOR

rrrr

PITTMAN 5111 WDGttl

Figure 3.1. Block Diagram of the Velocity Control System

28



DIRECTION OF ROTATION

N S N

ROTOR MAGNETS

HALL SENSORS

STATOR

WINDINGS

Figure 3.2. Position of the Hall Effect Sensors

5V

OV

5V

OV

SENSORS

FLUX

'

/ N \ / N \

\ 5 A f/ ZERO CROSSING

Time

SENSOR A

Time

SENSOR B

Figure 3.3

TIME

Flux in the Air Gap and Output VoltageWave Forms for Hall Effect Sensors.

29



The Hall sensor will produce square waves related to the

speed. Following from this- concept a digital tachometer

will be built and discussed in the next section.

The D input of the decoder in the motor drive is a

convenient logic input to apply a pulse width modulation

signal for speed and/or torque control. More details will

be discussed in later sections.

B. DESIGN OF THE DIGITAL TACHOMETER FOR SPEED CONTROL

The speed of the brushless DC motor can be observed from

the output of the Hall sensors. Hall sensors produce 2

square waves for each rotation. If elapsed time for each

revolution can be measured, the speed of the motor can be

found. One channel of the Hall sensor output of the brushless

DC motor is shown in Figure 3.4.

^The arrows Indicate the beginning and end of the period

of revolution. The relation between the period of the

revolution and the speed of the motor can be shown with the

following example:

Period = 1 Revolution = 50 10~3 sec

speed = 20 revolutions per second (RPS).

This is equal to 1200 revolutions per minute (RPM).

By starting from this approach, a digital tachometer was

designed by the author. The main idea was to measure the

period of revolution by using counters and inverting to the

voltage value by using a Digital to Analog converter (DAC).

30



£_ Mb/DlV oV/DiV

Figure 3.4. One Channel Output of Hall Sensor.

A circuit diagram of the digital tachometer is shown in

Fig., 3.5.

A 7474 Dual-D-Type positive-edge-triggered flip flop was

used to obtain 1 pulse per revolution by dividing the Hall

sensor signal by two. The output of the flip-flop is shown

in Figure 3.6.

74LS161 synchronous 4-bit counters were used to count for

each period. Clock pulses were used for the counters. For

this design the 16 bit procedure was found to be tne most

appropriate from a hardware point of view. When the motor

was running at a slow RPM, the period of the revolution was

high and the counter registered high. From an overflow

point of view, the maximum count on the counter should nj*

31



osc

600 KHZ

REGISTER (74374) _ REGISTER (74374)

>i

4X4 CASCADED COUNTERS(74LSI6I)

Figure 3.5. Circuit Diagram of Digital Tachometer.

32



PERIOD-

IC

<\-

_i

o>

T IME(SEC)

Figure 3.6. Output of the Flip Flop.

have exceeded 65536. Keeping in mind tnat wnen the motor

runs under 600 RPM the counter overflows, this criteria

became the minimum speed restriction for the motor. A

74121 monostable multivibrator was used to get short, clear

pulses for the counters. The output of the multivibrator

(one shot) is shown in Figure 3.7.

UJ

<\-

_i

o>

•PERIOD-

23NAN0SEC

T i ME(SEC)

Figure 3.7. Output of the Multivibrator

33



The 7^374 Register stored the counts for each period until

the new count came.

Two 8-bit DAC Digital to Analog converters were used to

convert the countsto

the voltage as it related to the

speed. The logic of the Digital to Analog conversion is

shown in Figure 3.3.

The voltage related with speed is between and 10

volts. When the speed is -40 rpm the output of the DAC will

be volts; when the speed is 24,000 rpm the output of tne

DAC will be 10 volts. The lowest speed is equal to 0, the

highest speed is equal to 10 volts.

V,_ CO TO 9.96 VOLTS)

0/A CONVERTER

MOST SIGNIFICANT 3ITS

9 TO 16 3ITS

0/A CONVERTER

LEAST SIGNIFICANT 3ITS

TO 7 9ITS

Figure 3.8. Digital to Analog Conversion.

C. PULSE WIDTH MODULATOR

The D input to the decoder is a convenient logic input

to which the pulse-width modulated logic signal can be

34



applied. It should be recognized that tine low mechanical

time constant of these motors could cause instantaneous

speed variation at slow speeds when a low duty cycle is

used in tne pulse width modulation. The pulse-width modulator

is shown in Figure 3.9. A pulse width modulated signal was

obtained by mixing a low frequency input error signal witn

a high frequency triangular dither signal. Twenty kilohertz

was the frequency chosen for the dither signal. The sum of

the error and triangular signal e(t) is shown in Figure 3.10A

T 2T

Figure 3.9. Pulse Width Modulator.

35



'dJ/tSt u\\J 5v'.' DlV

Figure 3.10. Error and Dither Signal.

An e(t) signal was fed to the zero crossing circuit. The

zero crossing circuit converts the resulting sum into a two

level signal e'(t) as shown in Figure 3.103. The signal

shifts between the two digital levels volts and 5 volts.

Input, eQ , is assumed to be a DC level or slowly cnanging

signal. Added to the triangular signal d(t), which oscillates

between -10 volts and volts, and has a period, T. This

signal was added to eQ

to produce e(t). This result was

then fed to a zero crossing detector, which in this case is

shown to switch from plus 5 volts (logic 1) to volts

(logic 0)

.

A circuit diagram of the pulse width modulator is shown

in Figure 3.11. A circuit diagram of the digital tachometer

36



and pulse width modulator is shown in Figure 3.12. The

artwork of the circuit is shown in Appendix C.

20KE o-A/W?—

20KD(t)0-A/W

ZERO CROSSING DETlCTOR

Figure 3.11. Circuit Diagram of the Pulse Width Modulator

37





IV. SYSTEM TESTING AND DATA COLLECTION FOR VELOCITY CONTROL,

A. GENERAL

After building the velocity control system for the

brushless DC motor some experiments were done to get data

on how the system works. The instruments used for these

experiments are shown below:

1. Power supply unit PS 150E

2. Hewlett-Packard 6216A power supply

3. .Wavetek model 145 pulse/function generator

4. Textronix 2213 oscilloscope

5. Textronix 464 storage oscilloscope6. Hewlett-Packard 3582A spectrum analyzer

7. Hewlett-Packard 85 plotter

8. Hewlett-Packard 124A camera.

The power requirements for the system were + 15V, -15V, +5V,

-10V, -15V.

DIP

SWITCH

(E>-

D/A CONVERTER

PWM

D/A CONVERTER —

DIGITAL TACHOMETER

(CK) (0A

Figure 4.1. 31ock Diagram of the Velocity Control System

39



For simplicity, test points were defined by letters. A

block diagram (and test points) of the system is shown in

Figure 4.1.

These test points are the same on the circuit board. For

velocity command, a four position dip switch was used. Fifteen

different speeds are produced depending on the relevant motor

power supply.

The calibration of tne system is important to getting

accurate data. For calibration purposes, many adjustable

resistors were used in the system. The calibration of the

system is explained in Appendix B.

B. OPEN LOOP VELOCITY CONTROL

For open loop studies, the feedback switch is turned to

the open loop (OL) position. The power supply was set to

15V. The Speed command was given by a dip switch. The

position of the dip switch and the equivalent RPM values

are as shown below:

Dip switch position 3peed (RPM)

0001 3000

0010 3260

0011 34100100 35700101 3660

0110 37500111 3300

1000 3845

40



The 3750 RPM speed (Dip switch = 0110) was chosen for the

first experiment. The two channel Hall sensor output of the

motor is shown in Figure 4.2. From the Hall sensor output

the speed of the motor can be calculated.

V/UIV

Figure 4.2. Hall Sensor Output of the Motor for 3750 RPM.

Since the Hall sensor sends 2 pulses per revolution, Figure

4.2 shows that

1 rev = 8 x 210 3 = 16 msec.

= 1/16 msec x 60 = 3750 RPM.

The pulse width modulated signal (test point P) is shown in

Figure 4.3.

When the shaft of the motor is held, the motor slows down

and no change of the pulse width modulated signal can be

seen. Another experiment was done by changing the power

supply of the motor. The speed of the motor was changed.

41



Figure 4.3. Pulse Width Modulated Signal.

Both these observations show that this is an open loop

system. In the second experiment 3260 RPM speed (dip

switch = 0010) was chosen. The Hall sensor and PWM signals

are,shown in Figure 4.4 and Figure 4.5 respectively.

1 iVIb/UIVp

Div

Figure 4.4. Hall Sensor Output of the Motor

42



Ô/Ji '

L I V Z V / U I

Figure 4.5. Pulse Width Modulated Signal for 3260 RPM.

C. CLOSED LOOP VELOCITY CONTROL

For the closed loop system, the feedback switch was turned

on to the closed loop position (CD. The power supply was

set to 30 V. The position of the dip switch and equivalent

RPM values are as snown below.

Dip switch position Speed (RPM)

0100

0101

01 10

0111

1275

1500

1580

1375

Due to hardware restrictions, a 16 bit system was used.

That brought some unwanted results in low speed experiments.

For that reason -2 V steady state error was added to the

Dither signal. The Dither signal is shown in Figure 4.6.

43



ZQ)i6iQiv ov/uiv

Figure 4.6. The Dither Signal.

For the first experiment on closed loop velocity control

the speed of 1275 RPM was chosen. The dip switch was set to

0100. The Hall sensor output of the motor is shown in Figure

4.7. From this picture the speed of the motor can be

calculated in the same fashion as the previous section.

Its speed is 1275 RPM. The PWM signal is shown in Figure

4.8. When the shaft of the motor was held slightly the PWM

signal was changed to keep up with the given speedcommand

(see Figure 4.9). This is one of the expected results of a

closed loop system. Another experiment was done by changing

the power supply of the motor. No change in the speed was

observed. This is another expected result of a closed loop

system.

44



b/pjy ?y/ uiy

Figure 4.7. Hall Sensor Output of the Motor for 1275 RPM

l_ 0/Jo ' UlV bV/ uiV

Figure 4.8. Pulse Width Modulated Signal for 1275 RPM

45



Figure 4.9.

D V ' U I V

Pulse Width Modulated Signal with ExternalForce on the Motor Shaft.

D. TRANSFER FUNCTION MEASUREMENT AND SIMULATION STUDIES

^The transfer function of the motor can be found by using

a spectrum analyzer. A Hewlett-Packard 3582A spectrum analyzer

was used for this experiment.

A block diagram of the closed loop velocity system and

its connections to the HP spectrum analyzer are shown in Figure

4. 10.

Random noise was used in the system and was fed to the

summing junction (test point N). When the forward gain of

the noise was 1.0, the speed of the system was changed due

to the noise. This unwanted result was eliminated by choosing

the noise gain equal to 0.2. The frequency response of

46



the system found by the HP spectrum analyzer as shown in

Figure 4.11a and Figure 4.11b.

MOTOR

test/POINTS

CIRCUIT BOARD

HP SPECTRUM ANALYZER

A B12 ?

NOISE

? Q Q

Figure 4.10. Closed Loop System with Spectrum Analyzer

X F R F C T N + 2 9 d B F S 1 9 d B - D I V

M K R - 2 3 . 2 d B

CD

Q

t -20

rj -30

<

,

r .

'M A,

1

v>il

i

rf

i

l III1

r

1

1

Lit

i

r 9

1 K R

Hz1 9 . 2 Hz Bl' 1 :

1

:5 _H2 i rnHz

I £.5

frequencyChz)

25

4.11a. Open Loop Frequency response of the systemwith magnitude curve.

47



LUQLU

<X

FRFR

F C T N »

F C T N

+ 8 9 d B F 3

M K R • 1 1 S o1 9 d B •

5 o /D IVd i v

LjI I

no

--

\r

uu

iftITT I rt i1

H-. 1 4 I 'I itfl

Ii

II

on jit jTlVJ VJ

rx I

\_

\

»

..

1KR

Hz

19.2 Hz

25 H

B U'

7 2 6 mHz1 2.5

frequency(hz)

25

Figure 4.11b. Open Loop Frequency Response of the Systemwith Phase Curve.

In the velocity control system there are a number of

various digital components, such as flip-flops, counters

and D/A converters. The counters which were used in the

system are synchronous devices, this means they use clock

pulses

.

The following events take place in the system.

1.

Wait for a clock pulse.

2. Determine the speed for one revolution of the motor.

3. Perform digital to analog conversion.

4. Send the updated control variable to the motor.

5

.

Go to step 1

.

48



Because the computation of the speed and sending the

control variable takes some time, there is a time delay

between steps two and four. The D/A converter holds the

signal over one revolution of the motor. This implies that

the sampling interval is equal to one revolution of the

motor. During the transfer function measurements, the

speed chosen was 1360 RPM. With simple calculation, one

revolution of the motor can be found to be 44 milliseconds.

The Nyquist frequency is thus /0.044 = 71.2 rad/sec or

11.3 hz.

At frequencies which are greater than the Nyquist

frequency, the ambiguities of the transfer function for

both the gain and phase curves can be seen in Figure 4.11a

and 4.11b. For that reason, this part of the experimental

data was not included in the calculations.

The frequency, magnitude and phase of the transfer

function which was found from Figure 4.11a and b are shown

in Table 1. The Bode plot which was drawn by using the

data in Table 1 is shown in Figure 4.12.

The transient response of the closed loop and open loop

system were found from a strip chart recorder and are shown

in Figure 4.15 and Figure 4.16. On the other hand, the

transient response of the system can also be observed from

the storage oscilloscope.

A Textronix 464 storage oscilloscope was used to get

the transient response of the system. The step input (from

49



1035 rpm to 1305 rpm) was applied to the system as a step

input. The closed loop transient response of the system is

shown in Figure 4.17. This transient response correlates

with the transient response which was found from the strip

chart (see Figure 4.16).

As can be seen, the system is type [Ref. 1] and has one

pole at w = 7.0 rad/sec and one pole at w = 27 rad/sec.

The open loop transfer function of the system is shown below.

GCa) = -^s/7.0+1 )(s/27.0+1 )

TABLE 4.1

FREQUENCY RESPONSE WITH MAGNITUDE AND PHASE

w(rad/sec) Magnitude(db) Phase(desrrees)

4 2.5 -91

5 2.6 -91

6. 2 2.6 -948. 7 0.7 -107

10 -0.9 -111

15 -3.3 -123

20 -6.8 -138

25 -9.2 -153

30 -11 .7 -164

35 -13.5 -176

40 -14.8 -19545 -17. 1 -207

50 -18.3 -212

55 -19.6 -23060 -22.2 -241

65 -24.9 -246

70 -28.9 -250

This transfer function was used for computer simulation

of the system. The open loop frequency response of the

50



system is shown in Figure 4.13 and the transient response

of the system is shown in Figure 4.14.

The time constant of the system was found from the open

loop transient response (see Figure 4.15). The time it

takes to get 63$ of velocity gives the time constant of the

of the system. From Figure 4.15 the time constant was

found to be 140 milliseconds. On the other hand, the

time constant of the system can be found from the transfer

function which was determined using the data from the HP

spectrum analyzer. The low frequency pole of the system as

determined from Figure 4.12 was 7.0 rad/sec, then the time

constant

1

T = = 142 milliseconds

7.0

This time constant correlates with the time constant which

was found from the strip chart recorder. This indicates thai:

the frequency response of the system which was found from the

HP spectrum analyzer was accurate.

The time constant of the closed loop system can be

calculated from the closed loop transient response of the

system which was shown in Figure 4.16. From the figure,

the settling time of the system was found to be 320 milli-

seconds. Thus the time constant of the closed loop system

was 320 milliseconds/4 = 80 milliseconds. It can be seen

51



that the time constant of the closed loop system was faster

than the open loop system. This was the expected result.

Another important subject arises from the usage of a

D/A converter in the system. Since the D/A converter

creates a delay related to the sampling rate, this will

cause phase lag in the system. This phase difference can

be seen by comparing the measured open loop frequency response

with that calculated from the transfer function. The

calculated phase does not include time delay, which the

measured phase does. It is seen that the measured phase

lag exceeds the calculated lag by 15° at the corner frequency

w = 7.0 rad/sec. Thus the time delay is approximately

<P

D= - 37.4 milliseconds

w

The time constant of the motor which was given by tne

factory specifications was 14.4 milliseconds. It is obvious

that the time constant of the motor is faster than the

system time constant. This difference is caused by the

time delay of the pulse width modulator and the D/A converter.

52



FREQUENCYCW,RAD/SEC)

Figure 4.12. Bode Plot of the System.

53



(2fap) asVHd0*581- (T08T-.

O-CT- 0'C2- 0'BB- <TB>-

(ap) aanxiNovw0-B3- 0*B»- O'Si-

Figure 4.13- Frequency Response of the System

from Computer Simulation.

54



1

oEs

1 j

i

1j 1

i i

12

i

i

i

:

i

:

:

i

j

j

:

:

—f j

1

i

i

j

|

•

1

j

j

r

r

-

—

: 1

1 i 1

i

1

1i 1 i ^r

a

T*T O'l 6*0 9-0 i'O 9-0 9*0 ro S'O ro ro o-o

C^da oooi v)a33ds indino

Figure 4.14. Transient Response of the System from

Computer Simulation.

55



H 1 1 1 1 1-

ggagSBts2̂ -H

-i 1i h

:v.¥.~ i:

:

-:M:TTfvr^—n——— 1o at-

^-j—i-

O ^;~Uh*t ...:

-:-:ii i^::r:

r^MMrSEeH h H 1 1 1-

Figure 4.15. Open Loop Transient Response of the Systemfrom the Strip Chart.

PRINTED »N U S *

-. t.— . : i

—^-£S-

^^

H 1-

t—r—

TVVT h—^_i._L

HP*f

2SBMZSBS It

-+-I

1 1 1i 1 1 (-

Figure 4.16. Closed Loop Transient Response of the Systemfrom the Strip Chart.

56



Figure 4.17. Closed Loop Transient Response of the Systemfrom Storage Oscilloscope.

57



V. POSITION CONTROL OF THE DC MOTOR WITH MICROPROCESSOR CONTROL

A. GENERAL

Microprocessor control of brushless DC motors has many

advantages over an analog control. One of the advantages is

that since it can be built with a couple of integrated

circuits, it is smaller and lighter than an analog controller.

It is also easy to debug the system.

There are some advantages and disadvantages to consider

in software design and its implementation as well. Some of

the advantages are:

1 ) By changing the software program, the function of the

system can be changed.

2) By. modifying the input/output devices, this system can

be used for other control systems.

3) By standardizing the hardware, system design emphasiscan be increased on software programs and subroutines.

4) Since the system is constructed of standardized units,

it is easy to debug the system.

B. MICROPROCESSOR CONTROL OF DC MOTORS

There are two approaches to microprocessor control. One

approach is the direct approach, another is the indirect

approach. In the direct approach the data obtained from the

system are fed into a microprocessor to compute the new value

of control. In the indirect method of microprocessor control,

the motor has an analog servo controller and microprocessor

58



is used to turn the servo on and off. {Ref. 2] In this

thesis the direct approach is used.

The block diagram of the microprocessor-controlled position

control system is shown in Figure 5.1 [Ref 33. The position

MICROPROCESSOR

CONTROLLER

n ii

PWM

DIRECTION

MOTOR

DRIVE

A OPTICAL

B ENCODER

a a<

r

0A 0B

HALL

SENSORMOTOR

Figure 5.1. Position Control System.

and direction commands are given from the cathode ray tube

(CRT) terminal. Another input to the microprocessor

controller is the actual direction of the motor which is

determined by using two channels of the optical encoder.

The direction sensing system is shown in Figure 5.2.

C. INCREMENTAL OPTICAL ENCODER

The incremental optical encoders are used for position

confirmation and for feedback signal generation. Incremental

optical encoders provide a pulse for each increment

59



of resolution. An incremental encoder has four main parts:

a light source, a rotation disk, a stationary mask, and a

sensor as shown in Figure 5.3. [Ref. 2] A Hewlett-Packard

Heds-6000 series incremental optical encoder was used for

the system.

DIRECTION

SENSORS

Figure 5.2. Direction Sensor.

LIGHTSOURCE

( LAMP, LED)

ROTATING

DISK

STATIONARY

MASK

SENSOR( PHOTOVOLTAIC CELL,

PHOTOTRANSISTOR,

PHOTODIODE )

Figure 5.3. Incremental Optical Encoder

60



The Heds-6000 series is a high resolution incremental

optical encoder. It consists of three parts: the encoder

body, a metal code wheel, and emitter and plate.

The incremental shaft encoder operates by translating the

rotation of a shaft into interruptions of a light beam which

provides output as electrical pulses.

The standard code wheel is a metal disc which has N=1000

equally spaced slits around its circumference. An aperture

with a matching pattern is positioned on the stationary phase

plate. The light beam is transmitted only when the slits in

the code wheel and the aperture line up. Therefore, during

a complete shaft revolution, there will be N=1000 alternating

light and dark periods. A molded lens beneath the phase

plate aperture collects the modulated light into a silicon

detector.

The encoder body contains the phase plate and the detection

elements for three channels. The first channel gives N=1000

pulses for each revolution. The second channel has a similar

configuration but the location of its aperture pair provides

an output which is in quadrature to the first channel. The

phase difference is 90° electrical. The direction of

rotation is determined by observing the leading form of the

channel 8. The outputs are TTL logic level signals.

The index channel is similar in optical and electrical

configuration to the A,B channel described above. An index

61



pulse of typically one cycle width is generated for each

rotation of the code wheel.

For counter clockwise and clockwise rotation of the code

wheel, channel A, channel B, and index channel outputs are

shown in Figure 5.4a and Figure 5.4b respectively. Encoding

characteristics, recommended operating conditions and

definitions are shown in Appendix E.

D. MICROCOMPUTER SYSTEM

The general block diagram of the microcomputer system is

shown in Figure 5.5. The microprocessor unit (MPU), Z-80,

implements the function of the central-processing unit (CPU)

within one chip. It includes an arithmetic-logical unit

(ALU), plus internal registers, and a control unit (CU), in

charge of sequencing the system. The Z-80 creates- three

buss.es: an 8-bit bidirectional data bus, a 16 bit unidirec-

tional address bus and a control bus.

The data bus carries the data being exchanged by the

different elements of the system. Mainly, it will carry data

from the memory to the Z-80 or from the Z-80 to an input/output

chip. The input/output chip is the component in charge of

communication with an external device.

The address bus carries an address generated by the Z-80

which will select one o. t be chips attached to the system.

For this system a 741S138 decoder was used.

This address specifies the source or the destination of

the data which will transit along the data bus.

62



UJQ

.CHANNELA

-r

CHANNEL B

CHANNEL

TIME OR ROTATION

Figure 5.4a. Encoder Channel Outputs for CW Rotation

LU

Q

_J

<

CHANNEL A

CHANNEL B

CHANNEL

TIME OR ROTATION

Figure 5.4b. Encoder Channel Outputs for CCW Rotation

63



osc

*

i

^> PORT A

^> PORT B

Figure 5.5. Microcomputer System.

The control bus carries the various synchronization signals

required by the system.

The Z-80 requires a precise timing reference which is

supplied by a 4.915 MHz crystal.

The RAM (random-access memory) is the read/write memory

for the system. MOSTEK MK 4118 (P/N) series, 1KX8 static

RAM was used for the microcomputer.

The system contained two interface cnips so that it could

communicate with the external world. The MC 68661B, Enhanced

Programmable Communications Interface (EPCI) was used to

communicate with the CRT terminal. The details on the EPCI

programming are explained in Appendix F. An Intel M8255A

64



Programmable Peripheral Interface (PPI) was used to interface

with the motor. The M8255A PPI has three ports which can be

used for input or output purposes. The operating modes of

the chip are explained in Appendix G.

The 2716 16K(2Kx8) UV Erasable Prom (EPROM) was used to

load the program for the system. The function of the

system can be changed entirely by writing the new program

and loading the EPROM. The circuit diagram of the microcom-

puter is shown in Figure 5.6.

E. SOFTWARE DESIGN

1 . General

The software was designed in such a fashion that a

position command to the motor is given from the CRT terminal.

The direction of the motor is calculated by the program which

chooses the CW or CCW direction for the shortest path to

its destination.

The system software was written in Assembly language

(Appendix H) at a Zenith Z-100 microcomputer, using a Z-80

instruction sets [Ref 31. The program was assembled and the

hex files downloaded to the EPROM by using a SYS19 routine.

The main program consists of:

1) an initialization routine for the ports and a CRT

interfacing,

2) calibration routine for a D/A converter, and

3) position control routine and subroutines.

65



2. Main program components.

The initialization routine sends a control word to

the parallel ports of the computer, setting them to the output

mode. Tnere are two options given to the user. First is

the calibration of the D/A converter (Appendix D) and second

is the position control of the system. After the calibration

of the system, the position command to the motor can be given

from the CRT terminal. For simplicity, the position of the

motor should be given as a count of pulses. Since the incre-

mental optical encoder gives 1000 pulses per revolution,1

pulse represents 0.36°. If the command is 100 counts, it

will represent 36°.

The direction of the motor is determined in the

following fashion. If the position command is greater than

180° (500 counts) the direction of the motor will be counter-

clockwise (CCW).

The program takes 300 states to calculate the position

of the motor and determine the new control command. The

actual time the program takes to execute can be found by

multiplying the number of states by the clock period. A

4.915 MHz clock was used for this microcomputer, so the

period of the pulse is:

1/4.915 106

= 0.2035 microsecond.

Each state would correspond to 0.2035 microseconds of real

time. By adding up the total number of states that the

program requires to execute and multiplying this by the clock

66



period, it can be determined how long this program will take

to execute.

300 states x 0.2035 microseconds = 60.6 microsecond.

On the other hand, the period of the pulses that are

sent from the incremental optical encoder should be longer

than 60.6 microseconds. Otherwise, the microcomputer will

miss the pulses and go to the wrong position.

At maximum, 810 rpm was found to be a sufficient speed

for the brushless DC motor. The motor will make one rotation

in 74 milliseconds and each encoder pulse period will be 74

microseconds long. This corresponds to 4 volt power supply

for the motor. When the position error is maximum, the motor

speed will be 810 rpm and it will decrease with a decreasing

error signal. When the error signal is between O -^ ' the

speed of the motor will be 600 rpm. The torque at this

speed was found sufficient to overcome friction in the motor.

The flow chart of the system is shown in Figure 5.7.

3. Description of the subroutines

To make the program useful and understandable some

subroutines were written.

The Getchar subroutine gets the character from CRT

terminal and stores it in register.

The Echo subroutine sends message string to the CRT

terminal

.

The Recall subroutine sends characters to the CRT

terminal .

67



The'

..

7

a i subroutine waits for the next positive rism,

edge of the encoder pulse.

The CPY88 subroutine calculates 6x3 bit mui tipiicaticn

in

i

-<t rt

m * r-+

nZ r- h- N. »o N.

IS—

'

IT

+^r «t

00 t

co m * of-

U.1 5

1roo-t

i O o .1 i N.o 2 2 _l _)

_ INI (0 * in :0

3 3 3 Z2 3 Z>

—''

' ''

'—I > '—I—1—1—I—I—I—1—I—I 1—I—I—

I

I I I I

O— (Mr^^m^r-oOcnO — <MKW IT) **>/> f\Jc0t^-<r>O rorOKjrorOw-itorOKlrorOr}- — — —

08Z °>z

WTT to

— N.(T>

I9IS1 91

tf . 01

%

r 1 2 u r o . o Circuit Diagram of the Microprocessor

63



INITIALIZE

PORTSSTA RT

GET POSITIONCOMMAND

X

DECREMENT

POSITION

CHECK ENCODERPULSE i

DIRECTION

/-SPEED-DIR=CW

INCREMENT

POSITION

GET ACTUAL

DIRECTION

'-SEPEED_DIR=CW

FIND POSITION

ERROR

GET ACTUAL

DIRECTION

FIND POSITION

ERROR

/_ SPEED

- DlR=CCW/

YES GET ACTUAL

DIRECTION

£INCREMENT

POSITION

.SPEED .

• DlR-CCW/ DECREMENT

POSITION

-SPEED

DIR=CCW/

GET ACTUALDIRECTION

INCREMENTPOSITION

'-SPEED

DIRzCW

DICREMENT

POSITION

SPEED-DIR=CW

Figure 5.7. Flow Chart of the Main Program.

69



VI. SYSTEM TESTING AND DATA COLLECTIONFOR POSITION CONTROL SYSTEM

A. GENERAL

A microprocessor controller using tne Z-80 was built

for the position control system. During the testing the

following equipment was used:

1. Power supply unit PS 150E

2. Hewlett-Packard 121 6 A power supply.

3. Wavetek model 145 pulse/function generator.

4. Power supply model 3650.5.

5. Hewlett-Packard 124A camera.

The power requirements for the microprocessor were

+15V, -15V, -10V, +15V and 3-30V. The power requirement for

motor drive as well as the incremental optical encoder was

+5V. A four volt power supply was used for the motor.

The sequence for turning on the power supplies for the

system is important. First, the power supply of the micro-

processor and motor drive should be turned on. The power

supply of the motor should be turned on at the very last.

The microprocessor system draws a total of 450 milliampers.

The maximum current limit of 500 milliampers should be set

before adjusting the five volt power supply.

To start the microprocessor the reset button should be

set. The dial which was mounted on the shaft to observe

the angular position of the motor can be adjusted to 0° as

an initial position.

70



B. SYSTEM CALIBRATION

The calibration of the system should be done before

using the system. For this purpose a calibration program

was written. After resetting the system, two options

appear on the CRT terminal. (See Figure 6.1)

After entering 1 for system calibration, a set of

instructions appear on the CRT Terminal. (See Figure 6.2)

The voltage on test point C should be adjusted to

-4.96 volts.

C. CLOSED LOOP POSITION CONTROL

After choosing the position control option from the

menu, a set of instructions appear on the CRT terminal.

(See Figure 6.3)

Since the optical incremental encoder has a resolution

of 1000, each pulse of the encoder represents 0.36°. The

position command should be given as counts. The relation

between counts and angular positions is given in Table 2.

A dial was used to easily observe the angular position of

the motor.

The block diagram of the position control system is

shown in Figure 6.4. The blow up picture of the curve

following block is shown in Figure 6.5. When the position

error is maximum the velocity will be 810 rpm. When the

position error is between minus 5° and plus 5° the velocity

will be 600 rpm. When the position error is minus, tne

direction of the motor is changed from the CW direction to

71



the CCVJ direction or from CCW direction to the CW direction,

depending on the initial direction of the motor.

The software program was written in such a way that

when the position error is zero the motor will not shop.

When the position error is 0.36° the direction of the motor

is changed to the other direction and position error is

-0.36° the motor is reversed again. This algorithm will

create a dither signal between +0.36° at the position.

This dither behavior will hold the motor shaft at tne

given position within +0.36°.

USE?

t-SKIHT

2-R^lliffliaHTSDL

Figure 6.1. CRT Terminal Menu for Program Options

72



•mi shutaîiiluE S

- IF YOU OH Kn fflJlKTHIWK [C Pflr

FWB AiOGiCfffr TIE RETURN

Figure 6.2. CRT Terminal Menu for Calibration of System

$ mnVM COHTraLflEOQWf t

-EMTIS IHb H1K111UH Iff OHWTS

- wixonr a» courtscj&a oekees)

- EHTER D» THREE DIGITS (812)

-HIT THE RETURN

Figure 6.3. CRT Terminal Menu for Position Control

73



TABLE 6.1

COUNTS AND ANGULAR POSITIONS

Counts Angular Position (degrees)

000142 15

083 30

125 45

167 60

208 75

250 90

375 135

500 180

625 225

750 270

875 315997 359

The software program was written in such a fashion that

when the position command was bigger than 180°, the program

would chose the shortest path for its destination.

Fifty runs for the position commands which were smaller

than 130° and fifty runs for the positions whicn were

greater than 180° were done.

For all the runs, the motor went: to the given position

and dither signal was found to be +0.48°. This was close

enough to +0.36° to be satisfactory.

The transient response and frequency response of the

system can be found by using the additional system interfacing

chips and by writing a new software program. This is

recommended for further studies in Chapter Seven.

74



MICROPROCESSOR

1

] VELOCITY

1y^-^ POSITION

ERRORî

/osition. ry~MCOMMAND* V-^y

PWMMOTOR

DRIVE1

^ -1

1 I

1

l_ i

MOTOR

J\OPTICAL

ENCODER

Figure 6.4. Block Diagram of the Position Control System.

POSITION ERROR(DEG)

Figure 6.5. Curve Following Block

75



VII. SUMMARY AND CONCLUSION

A. REMARKS AND CONCLUSIONS

The brushless DC motor has been shown to have some

advantages compare to the conventional DC motor. Brushless

DC motors with their disadvantages still are more favorable

for use in incremental motion applications. Since commutation

is done by switching transistors, pulse width modulation is

a desirable option in system design.

The low-cost position sensors such as Hall effect circuits

and optical sensing integrated circuits have been found to

be highly practical for servo designs. A velocity control

system designed by using the Hall effect sensors.

From the analyses, the time constant of the motor as

given in the factory specifications was considerably faster

than the measured time constant. This was the result of

the time delay of the pulse width modulator and the D/A

converter

.

The transfer function of the system was developed by

using an HP spectrum analyzer. The time constant of tne

system was found by using the transient response data which

was measured using a strip chart and storage oscilloscope.

The measure of the time constant was found to be identical

with the computer simulations of the system transfer function.

The position control of the brushless DC motor was studied

by using a Z-80 microprocessor controller. Position feedback

76



was obtained from an incremental optical encoder. The encoder

had 1000 resolution per revolution which provided high accuracy

for position control. Assembly language was used to write

a program for position control. For the Z-80 CPU a 4.915 MHz

clock was used. This brought the limitation for maximum

speed of the motor to 810 rpm.

The system testing for the position control system was

done and was found to be accurate. Since the incremental

encoder gives one pulse for 0.36° angular position, the

steady state error was programmed to be +0.36° to hold the

torque on the shaft. The steady state error whicn was

found from the position control system was +0.48°.

B. RECOMMENDATIONS FOR FURTHER STUDIES

For the digital tachometer a 16 bit (4x4 bit ) counter

system was used. By using the 24 bit counter system, the

performance of the system can be improved.

Eight bit D/A converters were used for both the velocity

and position control systems. By using 12 bit D/A converters,

the resolution of the system can be increased from 0.2

volts to 0.01 volts.

Instead of the Hall effect sensor, an incremental

optical encoder can be used with the velocity estimator to

measure the motor speed. The sampling rate will then be

faster than the sampling rate using Hall effect sensors.

77



A 2N 2222 transistor in the motor drive to which the

PWM signal is applied will burn out i-f the transistor

transistor logic (TTL) signal is used for the PWM signal.

To avoid this, the open collector logic signal with an

820 ohms pull-up resistor should be used for the PWM signal.

Assembly language was used to program the position

control system. There are many high level languages that

may be used such as Forth, Basic, Fortran, C, Pascal and

Ada. There are many advantages in using a high-level

language rather than assembly language because it takes

much less time to develop a system. The code is also much

more readable and therefore, easier to modify the program

with a high-level language.

The transfer function of the system can be found by

using a couple more parallel interfacing devices (Intel

8255A) and by modifying the program which was already

written .

Since the incremental encoder has two outputs with 90°

electrical phase difference, using botn outputs instead of

one output as a position sensor the steady state error can

be programmed to be +0.18°. This will require another CPU

with a faster clock.

It is recommended that after the circuits are built and

it is certain that it is working properly, it would be better

to build the circuit using wire-wrap technique to improve

78



the wire layout and also to reduce possible trouble shootini

errors .

79



APPENDIX A

RATING AND SPECIFICATIONS FOR PITTMAN 511 WDG #1

BRUSHLESS DC MOTOR

MOTOR PARAMETER UNITS SYMBOL VALUE

DAMPING CONSTANT (K K /Rj N.m/(rad/s)*D

1.42xl0~3

MOTOR CONSTANT (K / R) N.m/ W*M

37.7xl0~3

MECHANICAL TIME CONST.. cj/v ms T

M14.4

ELECTRICAL TIME CONST.

MOMENT OF INTERIA

. tvv ras

. 2

kg.m

TE

J

0.155

20.5xl0

6

VISCOUS DAMPING N.m/(rad/s) DF

13xl0~6

FRICTION TORQUE N.m TF

3.0xl0~

MOTOR MASS kg M 0.60

THERMAL TIME CONSTANT min TTH

15

THERMAL IMPEDENCE (WDG--AMBIENT) °C/W

°c

RTH

3.2

MAXIMUM WINDING TEMP.~MX

155

WINDINGPARAMETER UNITS SYMBOL VALUE

TORQUE CONSTANT N.m/A ** 29.9xl03

BECK EMF CONSTANT V/(rad/s) KE

29.9xl0~3

STATOR RESISTANCE ohms*T

0.631

STATOR INDUCTANCE mH L 0.0975

80



APPENDIX B

THE DAC CALIBRATION FOR VELOCITY CONTROL SYSTEM

The DAC system was set to to -10 volts output range.

If the system range is to be changed an adjustment in the gain

offset will be necessary.

To adjust the gain offset of the DAC the following

procedure snould be applied.

1) Turn off the power of the motor.

2) Turn on the power of the system.

3) Connect the test- point '0' to the ground.

4) Adjust the P1 pot until -5.00 volts is shown.

5) Adjust the P2 pot until -5.00 volts is shown.

81



APPENDIX C

ARTWORK FOR THE DIGITAL TACHOMETER AND PULSE WIDTH

MODULATOR CIRCUIT

niGiTai. TacHOfMTER . pwmNEIU n on

• o • fl

©• •

o o a a a a a oia«

10? 8tt,ftaof a aaaJ a aaa? aa aJ

o o) a a o a aojao>-«;

;J

J

r

ff

iota

*£

^

^s o o o o o a

• •a a

(Li

i-~A••• aoo

)«<<ti aooaoeoa oooooooa aoaisosoo

»

S-aaaaaa 1« JooooJo

0;

\±9J>fy

a •

^T *• a • a

82



*3

\\ trIio

Q

83



APPENDIX D

THE DAC CALIBRATION FOR POSITION CONTROL SYSTEM

The DAC system was set to to -10 volts output range.

Trie gain offset adjustment will be necessary for good system

performance

.

After pushing the start button, tne program will ask

to select an option for making calibrations. After selecting

the calibration option, tne microprocessor sends the signal

to the DAC. Minus 4.96 volts should oe seen from test point

'C'. If it is not, tne P1 pot should be adjusted.

34



APPENDIX E

CHARACTERISTIC OF THE OPTICAL ENCODER

Definitions

Electrical Degrees:

1 shaft rotation = 360 mechanical degrees= N electrical cycles

1 cycle = 360° electrical degrees.

Position Error:

The angular difference between the actual shaft position

and its position as calculated by counting the encoder's

cycles

.

Cycle Error:

An indication of cycle uniformity. The difference between

an observed shaft angle which gives rise to one electrical

cycle, and the nominal angular increment of 1/N of a revo-

lution.

Phase

The angle between the center of pulse A and the center of

pulse B.

Index Phase:

For counter-clockwise rotation is illustrated above, the

index phase is defined as

1 *2

, is the angle, in electrical degrees, between the falling

85



edge of I and falling edge of B. $ is the angle, in

electrical degrees, between the rising edge of A and the

rising edge of I.

Index Phase Error:

The Index Phase Error ( A$9

) describes the change in the

Index Pulse position after assembly with respect to the A

and B channels over the recommended operating conditions.

86



APPENDIX F

MC 68661B OPERATION AND PROGRAMMING

Prior to initiating data communications, the MC 68661B

operational mode must be programmed by performing write

operations to the mode and command registers. The EPCI can

be reconfigured at any time during the execution of the

program.

The MC 68661B register formats are summarized as follows:

MODE REGISTER 1 (MR 1)

MR 17 MR 16 MR1SMR

14 MR13 MR12MR

1 1 MR 10

Character Mode and Baud

Sync Asvnc Parity Type Parity Control Length Rate Factor

Async: Stop Bit Length

00 Invalid = Odd = Di'aplfd 00 = 5 bits 00 = Synchronous IX rate

01 = 1 stoo bit 1 * Even 1 = Enabled 0i=6 bits 01 = Asynchronous IX rate

10 « 1'i stoo oils 10 = T hits 10 = Asynchronous 16X rate

11 » 2 stoo oils n=8 bits 1 1 = Asynchronous 64X rate

Sync: Sync:

Number of Transparency

SYN char Control

Oouble » Normal

SYN t =• Transparent

1 Single

SYN

- MODE REGISTER 2 (MR2)

MR27 -MR24 MR23-MR20

T.C RiC P\n 9 Pin 25 T.C R.C Pin 9 Pin 25 Mode Baud Rata Selection

0000 E E T.C R»C 1000 E £ XbVNC R.C T-C Sync

0001 E 1 T.C IX 1001 E 1 T.C BKOET async

00 10 1 E IX R.C 10 10 1 E XSYNC R.C sync

0011 1 1 IX IX lOi 1 1 1 IX BKDET async See baud rates in labia l

0100 £ E T.C R.C i ioo E E xSmnC R.C T.C sync

0101 E 1 T«C 16X 1 101 E 1 T.C BKDET async

01 10 1 E 16X R.C 1 1 10 1 E xsrNC R.C sync

01 1 1 1 1 16X I6X 1 1 1

1

1 I I6X BKOET async

COMMAND REGISTER (CR)

CR7 CH6 CR5 CR4 CR3 CR2 CR1 CRO

Receive Transmit

Request Control Oat* Terminal Control

Operating Mod* To Sand Reset Error Sync Aiync iR>EN) Ready (T«EN>

00 = Normal rj^waiiun = Foicm RTS » Normal Async

Ol » ASyDC output n.ijn 1 » Heiel Force oreah

Automatic one ciork iimo error flags Normal * Onarjtt) » Force OTR » Disable

ecno rm.tl* ditbf r*brt n sl«fu<. rmjister t =*Foir.e fwea* i - Endtiie output riign i Enable

Sync SYN .,nx3 or i«i ii./j'-i (FE Ofc PE OLE i * Force OTR

Ol fc stripping moo» i » Force «TS uulot rl OUtPut luaat

i0 Local i-ioo r>acn Output low

t 1 = flemoie in <u oa^h

Sync.

Send OLE

u a Normal

i » Sena OLE

87



There is one MC 6866 1 B device in the system. Mode register

1 address is CE Hex, mode register 2 address is 7D Hex and

command register address is 5 Hex.

88



APPENDIX G

INTEL 8255A OPERATION AND PROGRAMMING

The Intel 8255A contains three 8-bit ports (A, B, and

C). All can be configured in a wide variety of functional

characteristics by the system software. There are three basic

modes of operation that can be selected:

Mode - Basic Input/Output

Mode 1 - Strobed Input/Output

Mode 2 - Bi-directional Bus

Mode definition control word format is as follows:

CONTROL WORD

D2

ID,

GROUP 8

PORT C (LOWER)

I - INPUT- OUTPUT

PORT 3

I - INPUT- OUTPUT

MODE SELECTION

0- MODE

I _ MODE I

/ GROUP A

PORT C(UPPER)

I - INPUT- OUTPUT

PORT A

I - INPUT

0- OUTPUT

MODE SELECTION

00 -MODEI -MODE I

02-MODE 2

MODE SET FLAG

I -ACTIVE

89



There is one 82557A device in the system. Port address

is 39 Hex. This means port A and port 3 are at output; port

C is at input mode.

90



APPENDIX H

MAIN PROGRAM

POSITION CONTROL S. CALIBRATION PROGRAM

PPIA EOJ 1522.1

FFI3 rCU 1 32 1

PPIC EOJ 1S2 2P.

FPICONT ZCJ 1=23?

EDATA ECJ 1222P.E S T A T EOJ 1ZZ1H

emote fou 1Z22HECOMD ECJ 1223F.

RAM BASF EOJ 3223

PCS ECU SZ1HDIR ec: 32 3 P

DIRE ECJ 32 1i

CHAR ECU 5ZSHCOUNT ECJ 325B

I PEAI ECJ SZ7H

MFRAD ECU 309 H

RES AD ECJ 32EH

SUM1 ECJ S1ZH

SDM2 ECU 5125

CUM 3 ECJ 514HML1 EOJ 51 ~H

ML2 ECU 317H

M.L3 ECJ 31 5 P.

VEL ECU 51 3R

CR V ^ T 2DP.

ECU 2 AH

CP.3 2222

LD ?P , 2EFEH

LD A.ZCEELD (zmode: ,a

LD A .7DR

LD (CMor r > a

1 j_ i J L- L. f -T.

LD A,

LD ( TQQ^^ > ^

LD A.S9HLD 'PPICONT ,A

L001 : LD A ,2

LD ( FP I A , A

LD IX.HEAD1CALL ECHO

LD I X . £ P A C E

CALL ECHC

LD IX.HEAD2:all ECHOrn

IX.HEAD3CALL ECHO

LCQ11 : CALL C-E7CHAF.

SEI STACK POINTER

SET *ODEl P.EC-ISTE?. FOR EPCI

SET MODE? REGISTER FOR EPCI

SET COMMAND REGISTER FOR EFCI

SET MODE REGISTER FOP. PPI

[A ° ACT E Dro

C1

91



L002:

START1

l: i x , c a a ?r ALL RECALL

LD A . (CHAR)

SEC *.3EHC? 1

J? Z.LCC2

LE A, (CHAR)

SBC A, 523

CP 2

JF Z.STARTl

CALL ERROR

JP LCOll

LD IX. HEADSCALL ECHO

LD IX, SPACECALL ECHO

Lr IX.5EAD5CALL ECHO

LE IX.HEAD7CALL ECHO

LC IX, HEADSCALL ECHO

LC IX, HEADSCALL ECHO

LD A . 7 EH

LE (ppia; ,a

CALL GETCHARCP :r

JP Z.LOOl

LE IX, ERRORCALL ECHOJ? L002

LE IX.EEAD12 1

CALL ECHO

LE IX.HEAE11

CALL ECHCLE IX.HEAE12CALL ECHC

LD IX.EEAD12CALL ECHO

LE IX.HEAD14CALL ECHO

CALL CETCHARLD A, (CHAR)

LE IX. CHARCALL RECALL

SBC A ,3ZH

LD (KL1) ,A

CALL GETCHARLD A, (CHAR)

LD IX. CHARCALL RECALLSBC A.32H

LD (ML2) ,A

CALL GETCHAR

; CALEBRAIION PF.D3PJW

; FCSITIC'I CONTROL PP.CSHAK

J CALIBRAIION PROGRAM

; SEND CALIBRATION SIGNAL SEND TO THE PORT

; IS IT CARRIAGE ?

I PRINT HEADER

IGET POSITION FROM CRT

? POSITION (ASCII : > A

STRIP ASCIIFIRST DIGIT

GET POSITION FROM CRTPOSITION (ASCII ) > A

STRIF ASCII

SECOND DIGIT

GET POSITION E OM CRT

92





ANSC :

START2:

CCW2:

C W2 :

C0.NT2A

P0SIT2:

CV2A:

P0S1A

SEP1:

CONT1

WAIT1

Lr A.(DIR)Lr (ppib;,a

LC A.S2ELD (PPIA),A

LD A.(DIR)AND Z1HJF Z.STAP.T2

JP STRTZA

LD A,(PPIC)LE 3,

AND 3ie

JR Z.START2LD A,HAND 223

JP Z.CW2LD A,l

LD (EIRE) ,A

JP C0NT2A

LD A.Z

LD (EIRE),

LD HL,(F0S)AND A

SBC HL.BC

JP Z.ME5AT2JP M..NEGAT2

JP P.POSIT2

LD A, (DIRE)AND 213

JP NZ.CCW2AIMC EC

LD 3L, (PCS)

AND A

SEC EL, ECJP P ,P0S1ALD A, 3

CFL

LC 3.

LD A.L

CPL

ADD A,l

LD L.A

LE D 3

LD i'.i

LD L.13ELD 3.2

SEC 5L.DEJF M.SEP1LD A.2D5BJR CONT1LD A ,ZES5

LD 3,2LD (PFIA),ALD A,

LE (PPIE).ALD A.(PPIC)

; c*=? ccw=i; DIRECTION > CV

; direction > ccv:

j check the sncoeer; a — > 3

nc pulsf check again

P3ASE B > A

C*'=2 CCV = 1

SET POSITION

COMPARE THE POSITIONAT THE POINTBETOND T3E POINTNOT AT THE POINT

DIRECTION OF THE MOTOR

CCtf = l CV.'=2

CLOCK VISE ROTATIONLOAD POSITIONCLEAR FLAGS

COMPARE THE POSITIN

; CCMFLE M EMT

; COMPLEMENT

.76 de: POSITION LIMIT

; SPEED COMMAND

; SEND SPEFD COMMAND

; SFNB DIRECTION

; CHECK THE ENCODER

94



CCW2A

P0S2A:

SFP2:C0NT2

WAIT2:

NEGAT2:

CV2B:

P0S3A:

AND 31HJF NZ.VAIT1JP STAP.T2

DEC EC

LD 5L, (POS)

AND A

SEC 3L,BCJP P .PDS2A

LD A ,a

CPLLD a, a

LD A,L

CPL

ADD A ,1LD L.A

LD D,aLD E.L

LD L.1ZH

LD a,

SEC HL.DE

J? M.SEP2LD A.2DSHJR C0NT2

LD A.ZESHED a,

LD (PPIA) ,A

LD A.HLD (PPIB).A

LD A.(PPIC)AND Z1EJ? NZ.WAIT2JP STARTS

LD A.(DIRD)AND 21H

JF NZ.CCW2BINC BC

LD HL, (F3S)

AND A

SBC HL.ECJP P.P0S3ALD A,H

CPL

LD H,A

LD A,L

CPLADD A .1

LD L.A

LD D,E

LD E.L

LD L.12H

LD H, 2

SBC HL.DE

JP M ,SEP3

LD A,2D2H

JR CCMT3

5 COJNTERCLOC WISE ROTATION

J LOAD POSITION

J CLEAR FLAGS', COMPARE THE POSITIN

J COMPLEMENT

J COMPLEMENT

; 5.76 DEC. POSITION LI W IT

i

; SPEED COMMAND

; ctf=e

; SEND SPEED COMMAND

', SEND DIRECTION

;

CHECK THE ENCODER

J CCW=1 CV.=2

J CLOCK WISE ROTATION

J LOAD POSITION; CLEAR FLAGS

J COMPARE THE POSITIN

J COMPLEMENT

J COMPLEMENT

5.76 DEC. POSITION LIMIT

I SPEED COMMAND

95



SZP3 :

CCNT3

V-AIT3:

CCV2B

P0S4A

SEP4:C0NT4:

WAIT4:

STRT2A

STARTS:

CCW3:

CW3

LC A. .ZEES

LD 2.1

le (FPU), A

LD A.E

LE (PPIB) ,A

LD A.(PPIC )

AND Z1HJP NZ.WAIT2

JP STARTSDEC EC

LD HL.(PDS)

AND A

SEC HL.BCJP P .P0S4A

LE A ,fl

CFL

LD H.A

LD A,L

CPLADD A,l

LE L,A

LE D ,H

LE E,I

LE L.10HLE a.

SEC HL.DPJP M.SEP4LD A.2DSHJR CQNT4

LE A.2EEHLB 3,1LE (PPIA).A

LD A,

LE (PPIE) ,A

LD A.(PPIC)A 'E Z1EJP M7.WAIT4JP STARTS

LE DE. (PCS)

LE HL.Z3E7EAND A

SEC 3L.DELE (POS) ,HL

LD A.(PPIC)

LE H.AAND 215J P. Z, STARTSLD A,H

AND Z2KJP Z , C V/3

LE A.lLE (EIRD).AJP C0NT3A

LD A, 2

LD (DIRD),A

; :r*=i

i SEND SPEEE COM1ANE

J SEND DIRECTION; CHECK TEE ENCODE?.

; CCUNTERCLCCK VISE ROTATION

; LOAD POSITION

I CLEAR FLAGS

; COMPARE THE POSITIM

; COMPLEMENT

; COMPLEMENT

; 5.75 DEC. POSITION LIMIT

J SPEED COMMAND

J CCrf=l

; SEND SPEED COM1ANE

SEND DIRECTION

; CHECK THE ENCODER

; 192 DEC. LIMIT

; SHORTEST PATH; CHECK THE ENCODER

J NO PJLSE CHEC AGAIN

; cv=? ccw=i

96



LD 3L,(FCS)AMD A

S EC P.L.EC

JP° Z .NEGA.T3

JP P.P0SIT3JP M.NZGAT3

P0SIT3: LD A.fDIRD)AND aiH

JF NZ.CCW3ACW3A: DEC BC

LD HL.(PDS)

AND A

SBC HL.EC

JP P.FCS5ALD A ,H

CPL

LD E,ALD A,L

CPLADD A .1

LD L,A

P0S5A: LD D,H

LD E ,L

LD L.12ELD a, 2

SBC HL.DE

JP M.SEP5LD A.2DSB

JR C0NT5SEP5: LD A ,ZESH

C0NT5: LD H.l

LD (PPIA),A

LD A,HLD (PPIE),A

WAITS: LD A.(PPIC)AND 21HJP NZ, WAITSJP START3

CCV3A: INC BC

LD HL.(POS)AND A

SBC HL.EC

JF P.PCS5A

LD A.HCPLLD H.A

LD A,L

CPL

ADD A.l

LD L,A

P0S5A: LD D,HLD E.L

LD L.1ZS

LD a,

SBC HL.DE

; GET FCSITIG.

; ccw=i cv=?-, CLOCK WISE ROTATION

J LOAD POSITION

J CLEAR FLAGS

; COMPARE THE POSITIN

J COMPLEMENT

J COMPLEMENT

J 5.75 DEG. POSITION LIMIT

J SPEED COMMAND

; CCW = 1

J SEND SPEED COMMAND

; SEND DIRECTION

J CHEC? THE ENCODER

ICOUNTERCLOCK WISE ROTATIONJ LOAD POSITIONJ CLEAR FLAGS

J COMPARE THE POSITIN

J COMPLEMENT

J COMPLEMENT

', 5.75 DEG. POSITION LIMIT

97





pose a

SEP 9:

C0NT3:

WAIT8:

LD L,ALE D,H

LD E,LLD L.1ZHLD 3.2SBC HL.DF

JP M .SEP^LC A.2DSHJR CONTS

LD A.ZESHLD B,2

LC (PPIA).A

LD A.HLD (PPIB),ALD A , (PPIC >

AMD ziaJP NZ. WAITSJP START3

; 5.75 DEO. POSITION LIMI

'

; SPEED COMMAND

; cw=e

; SEND SPEED COMMAND

J SEND DIRECTION

; CHECK THE ENCODER

HEAD1HEAD2:

head:HEAD4:

HEADS:

HEAD5:

HEAD7:HEADS:

FEAD9:HEAD12:

HEAD11:

HEAD12:

HEAD13:HEAD14:

SPACE:QUEl:

0UE2A:

3UE2E:SONM :

ERROR:

COUN:Rl:

R2:

DB

DE

D3DB

DE

DB

DBDE

DEDB

DE

DB

DE

DE

DBDB

DB

DE

DBDE

DE

DE

DB

CR.LF,

'CR.LF,

CR.LF.CR.LF,

CR.LF,

CR.LF,

CR,LF.CR.LF.

CR.LF,CR.LF,

CR.LF,

CR.LF,CR.LF,CR.LF.

CR.LF,CR.LF,

CR.LF.

CR.LF,

CR.LF,CR.LF,

CR.LF,CR.LF,CR.LF,

* WHICH1-

2-

EMTER* SYSTEM

- CHEC- you- if r

-IF Y* PCSITI

- ENTE- MAXI- ENTE- HIT

. '.CR.LFENTER TH

ENTER T3

CV = 2

MOTOR AT

E^ROR ?

+++ POSI

READY TO

COUNT HA

PRO

SYS

PCS

THE

CA

i r

S50

CU

CU*M

<p

vlfTM

R I

.'$

1 P

E D

ccv;

TH

T

TIO

CRAM MTEM CAL

ITION C

NUMBEP

LI BRAT I

HE CHEC

ULD SEEDC NOT,

ARF DONCONTROLHE POSI

939 CO

N THREERETURN

ULD YOUIERATIO^

CNTROL'

AND HITON PROG?.

EC POINT-4.35 7

ADJUST

F HIT THPROGRAMTION IN

UNTS ( 3

DIGITS', CR.LF

LIKE TO USE ?' CP ,LF,'4

',CR.LF,'$'

.CE.LF.'i'THF RETURN ', CR, IT,

'4

'

AM *',CR,LF.'$*

C '.CP.LF,'?'

OLTS '.CR.LF,' %'

WITH 52 £ POT .', CR.LF'

E RETURN'

,CR,*',CR,LF,VLF,

'%'

COUNTS '.CR.LF /%*

59.4 DEGREES ) CR .LF,'4T

(212 N' '.CR.LF, '4 '

'4 '

OSITION IN C0UNTS'292 1 ' ,C?.,L

IRECTION '.CR.LF, '%'

= 1 ', Z P. , LI ,

'$

'

E JIVEN POSITION ' CR.LF.TRY AGAIN '.CR.LF, 'S'

N + + +',CR, LF, '$'

ND COUNT' .CR.LF. '$'

ENT' .CR.LF. '%'

F.'$

99



subroutines

HTCHAR: LD A . (ESTAT)

AND 2

JR Z.GETCHAR

LD A , (EDATA X

LD (CHAR),

RET

;]ET EPCI STATUS

J IS A CHARACTER ENTERED ?

;\'C, CHECK AGAIN

ITES , C-ET CHARACTER; STORE IN A

ECHO

FIN

ir A,(ESTAT JGET

AND 1 ;is

JR Z.ECH3 .NO,

LC A, (IX) ;lca

CP '%' ;che

J P. Z.FIN ;las

LD (EDATA) ,A ;sen

INC IX ;nex

JP.

RET

ECHO ;xmi

EFCI STATUSEPCI READY ?

CHECK AGAIN

D MESSAGECK THE LAST CHARACTERT CHARACTERD CHARACTERT CHARACTERT NEXT CHARACTER

RECALL: LD A, < ESTAT) JC-ET EPCI STATUSAND 1 JIS EPCI READY ?

JR Z, RECALL ;W0, CHECK AGAIN

LD A, (IX) J LOAD CHARACTERLD ( EDATA), A J SEND CHARACTERRET

CPY83:

MULr

NOADD

LD BC , (YPRAD)

LD E,SLD DE. (MPDAD)

LD D,3

LD HL.2

SRL C

JR NC, NOADD

ADD HL.DESLA E

RL D

DFC B

JP NZ.MULTLD (RESAD) ,HL

RET

DS 22

5ND

100



LIST OF REFERENCES

1. Thaler, J. G. Design of Feedback Systems , Dowden,Hutchingson & Ross, Inc., 1973.

2. Kuo, B. C. and Tal , J. DC Motors and Co ntrol Systems .

SRL Publishing Company, 1978.

3. Zaks, R. Programming the Z-80. Sybex Inc, 1982.

4. Sears, F. W. and Zemansky, M. W., University Phvsics,

Addison-Wesley Publishing Company, Inc., 1964.

101



INITIAL DISTRIBUTION LIST

2. Library, Code 0142Naval Postgraduate SchoolMonterey, California 93943-5000

3. Professor Harriett Rigas, ChairwomanDepartment of Electrical and

Computer EngineeringCode 62RrNaval Postgraduate SchoolMonterey, California 93943-5000

4. Professor George J. ThalerCode 62Tr

Naval Postgraduate SchoolMonterey, California 93943-5000

5. Professor Alex Gerba Jr.

Code 62GeNaval Postgraduate SchoolMonterey, California 93943-5000

6. Istanbul Teknik Universi t iesi

jilektrik Fakdltesi

GdmQssuyu/ IstanbulTURKEY

7. Deniz Harb Okulu KomutanligiOkul KQtQphanesi ve ElektrikBSiama KQtaphanesiTuzla/IstanbulTURKEY

8. Mr. J. C. Cochrane, M.S.

665 W. Lake Hazel

Meridian, Idaho 83642

9. Mr. Nusret YUrdtUcU, M.S.

7220 Trenton Place

Gilroy, CA 95020

10. LTJG Nezih DurusuMenpare sokakNo. 1 HeybeliadaIstanbul TURKEY

102

No. Copies

1 . Defense Technical Information Center 2

Cameron StationAlexandria, Virginia 22304-6145







DTr LIBRARY,,DUATE SCHOOL

A 93945-6003

Thesis

D906

c.l

219517Durusu

Brushless DC motors,velocity and positioncontrol of the brushlessDC motor.

Thesis

D906

c.l

213517

Durusu

Brushless DC motors,

velocity and positioncontrol of the brushlessDC motor.



Brushless Dc Motor 00 Du Ru

Documents